Face image complementing method based on self-attention deep generative adversarial network

Abstract

The invention provides a face image completion method based on a self-attention deep generative adversarial network. The method comprises the following steps: constructing a face image completion model comprising an attention recurrent neural network module, a generator network and a discriminator network; carrying out data collection: collecting a large number of face images to form an image set, and dividing the image set into a training set and a test set; carrying out image preprocessing to enable the size of the image to be suitable for processing in a deep learning network; constructing a damaged face image serving as model input; carrying out model training, using a GAN framework and various regularization means to directly train the generator network and the discriminator network at the same time in an end-to-end mode, and when the generator network and the discriminator network reach theoretical Nash equilibrium, completing model training; training the model. Compared with the prior art, the face image complementing method based on the self-attention deep generative adversarial network has the advantage that the complemented face image is better in quality.

Description

technical field [0001] The present invention relates to the field of image processing and deep learning, in particular to a face image completion method based on self-attention deep generative confrontation network. Background technique [0002] When digital images are transmitted over the network, stored on the hard disk, or taken improperly, it is easy to cause incomplete, partially incomplete or damaged images. Therefore, image completion technology can be widely used in the restoration of cultural relics, restoration of damaged or partially polluted photos, film and television Special effects production, text in image, watermark, obstacle or specific object culling and so on. However, the current mainstream image repair solutions only use traditional methods such as image search, matching, and filtering to complete the completion. The core of this type of solution is to use the redundancy of the image itself, based on patch-based image synthesis, and has been successfull...

AI Technical Summary

Problems solved by technology

[0009] Aiming at the technical problems of inability to perceive global semantics and capture long-distance spatial information in related technologies, resulting in poor quality of patched face images

Images